Only three weeks left before SYCL 2024 in Milan! And it’s packed with great talks and workshops: sycl.it/agenda/

Finding the (apparent) lack of SYCL Mac Silicon GPU support disheartening along with the realization that to Intel it's OneX86API. Same as it ever was.

SimSYCL: A SYCL Implementation Targeting Development, Debugging, Simulation and Conformance

#SYCL #HPC #Package #IWOCL

hgpu.org/?p=29190

Want to increase the performance of your C++ application? Learn how you can use the Intel #oneAPI DPC++ Library to add advanced parallelism to your application with #SYCL for multiarchitecture programming: intel.ly/44lpbTH

The 2024.1 Intel® Software Development Tools are now available and include a major milestone for accelerated computing:
Intel® oneAPI DPC++/C++ Compiler has become the first compiler to adopt the full SYCL 2020 specification.

intel.com/content/www/us…

Need to launch many offload kernels while using resources efficiently and keeping overhead to a minimum? Try the #SYCL Graph preview feature in Intel #oneAPI DPC++/C++ Compiler 2024.1. #UXL
intel.com/content/www/us…

Sasank Chilamkurthy Got Cinny running against it for a bit, issues, will try again later. In meantime, thx for SYCL Academy combined slides (outline.von-neumann.ai/s/d0cd5eb9-2e1…). We need to target Mac, looking into SYCL robustness there next. Also, some concerns about Guile on Windows, but seems tractable.

Dive into the world of accelerated programming with C++! 🚀 Learn to harness data parallelism and SYCL in this updated second edition. Get ready to supercharge your C++ programs! #Programming #Cplusplus #SYCL Intel #openaccess

📚 Download ebook here: link.springer.com/book/10.1007/9…

SYCLさん行ってきた！
ずっとオンラインでは拝見してたものの、やっぱり現地はすごい！
ずっと、「はー！」とか「へー！」とか言うてた笑
めっちゃ勉強になりました◎

かげさん( かげ┃コミュニティマネージャー )ご案内ありがとうございましたー！！

You haz attention. Read your slides (twitter.com/sasank51/statu…). Github? No pressure. What Lisp? Publicly committing to Clojure insanity ATM, but dig SYCL & walter lynsdale may forgive me if I release some code and give you cycles. Relevant AAPL, AMD, INTC & Tenstorrent HW at hand.

Horace He Aart Bik LLVM was a backend for C++. And that proved generic enough for other programming languages and architectures. We need something similar for GPUs. Triton is a tad bit too high level. A backend targeting SYCL might be low level enough.

Want to know why and how AdaptiveCpp outperforms vendor compilers (nvc++, roc-stdpar, icpx) for C++ parallel STL offload? Read my brand-new IWOCL paper: dl.acm.org/doi/10.1145/36…

Horace He Aart Bik SYCL works on Nvidia, Intel and AMD as well. But I guess there may have been a reason why there is `torch.cuda` and not `torch.gpu`, even for AMD.

I need to find C and Cpp community 🥲.

A few months ago I’d just be chilling with different flavours of JavaScript, occasional Python, and doing some tech lead stuff.

Now I’m involved with C, Cpp, OneAPI, sycl, etc.

And who is Grace Hopper anyway? 🤷🏽‍♂️
She keeps coming up

小医生……是伯恩山，非常的可靠稳重，看到就给人一种安心感呢……
本篇毫无疑问是德牧，看起来很帅气，忠诚护主大狗狗
哈迪斯，嗯……感觉是狼犬吧，看上去很可怕其实只是不会表达

idk dude. support SYCL? Optimize Vulkan? actually make good OpenCL?

Researchers at Indian Institute of Technology Goa achieved a 1.9x performance improvement on their 2D Poisson Equation Solver by migrating their code from CUDA to #SYCL using tools from the Intel #oneAPI Base toolkit. Learn how: intel.ly/49JPKmu

#HPC #GPU SYCLstd

Learn how to implement the #Viterbi #algorithm (a dynamic #programming approach) to solve Hidden Markov Models using #SYCL . Leverage the Intel #oneAPI DPC++/C++ #Compiler for offloading complex computations to #GPU .
intel.com/content/www/us…

#HMM #markov #garph #cpp #WeAreIntel

Jon Masters 🏴‍☠️ I’ve been hoping for a while that “natively parallel” languages (SYCL amongst others) and high level synthesis would be a means of opening that up.

The toolchains are getting better by FPGA standards?